system analyst and adjunct cio business analyst and technical writer

 Home  Who We Help  Services  Approach  Case Studies  Resources  Contacts  About Us


If your company is like most, you have many software packages in use throughout the operation. Finance has their accounting package. Sales has CRM. Production has ERP. HR has performance management. Everybody's happy...until they need to share information with another group. That's when they fall back on email or sneaker-net.

You've spent hundreds of thousands, perhaps millions, of dollars on software, yet your departments still rely on manual efforts to share information. The left hand doesn't know what the right hand is doing. Here are two examples of this phenomenon.

Example 1

A major systems integrator and value-added reseller purchased Goldmine for it's Sales and Marketing departments. It also purchased Microsoft Great Plains for its Finance department. Now what happens? The sales and marketing folks manually enter prospect and customer information into Goldmine. The finance folks manually enter the same information into Great Plains.

This immediately causes two problems. First, data quality issues; mistyped information resulting in customers that appear to exist in one system but not the other. Second, differing views of customers; sales thinks CompanyX is a high-value customer. Finance thinks CompanyX is a terrible customer because they have to be hounded for weeks to get payment.

They need to get on the same page!

Example 2

A well-known retailer of gift items invested a large sum in a custom web-based, e-commerce site. They also purchased Avexxis software to handle all back office processing including order fulfillment. All the data collected by the website is continuously transmitted to the Avexxis software.

They are very good at filling orders. But there's a major problem. The website doesn't maintain any information about their customers. Even simple questions can't be answered. Is this a return visit? Has the person bought before? What did they buy? The situation is complicated by the fact that customers can also order by telephone.

As a result, opportunity is lost. No up-sell. No cross-sell. No incremental revenue. Sales are flat. They have the data. They need to leverage it!

What do they do now?

The problem is not in the software. Goldmine, Great Plains and Avexxis provide good software products. The problem is in how the companies that buy these products fail to take full advantage of what they have.

If you describe these problems to the major data integration vendors (eg. Microsoft, Oracle, Business Objects, Cognos, etc.), you'll hear about the need to extract, cleanse and aggregate data. They'll go on for hours describing how they analyze, clean and reformat data.

It sounds good. It's the wrong approach!

The cost-effective solution lies in preventing errors not fixing them. Manufacturing firms work relentlessly to prevent mistakes from happening. Why? Because it's far cheaper than hiring an army of "fix it folks" to correct mistakes at the end of the production process. That's the 1000 points of light solution. You need a high energy laser!

Think Lean

Ever heard of "Lean Production"? It's a reference to a set of techniques used by manufacturers to minimize cost and maximize value. (See the side bar for a summary.)

You need to apply these techniques to information management. Here's how. Let's walk through the ten rules of Lean Production as they would apply to managing typical business content.

1. Eliminate Waste - Analyze what information is captured and how it is saved. Flag redundant activities with the goal of entering information once.

2. Minimize Inventory - What inventory? Think of the information you collect as inventory. Only save what is useful and only save it once.

3. Maximize Production Flow - Business information does not age well. It needs to be made available as soon as possible. Don't try to gather all the information you need before making any of it available.

4. Pull Production from Customer Demand - Okay, but who is the customer? In these situations, the "customer" is the sales department or the finance group or whoever needs the information. Allow them to control the information and its usage. Be sensitive to their needs.

5. Meet Customer Requirements - Be sure you fully understand what expectations each group has of the information being collected. What data fields do they need? Don't overlook the simple. For example, is zip code 5 digits or 9?

6. Do It Right The First Time - Simple error checks can be built in at the point of data entry to capture the most common mistakes. Don't allow the database to be filled with bad data. It frustrates everyone!

7. Empower Workers - Allow people to flag and/or fix data errors when they find them. Provide feedback to the source of the error so it can trapped the next time.

8. Design for Rapid Changeover - Software upgrades and changes are ever present. Keep the process generic and minimize the custom software development.

9. Partner With Suppliers - If the software you're using doesn't do what you want, make noise! Be heard! Put it in writing! They will listen.

10. Create a Continuous Improvement Culture - Explore. Plan. Execute. Verify. Then do it again. And again. Don't ever stop.

Simple Solutions Work Best

Let's return to the examples are the beginning of this article. What should the system integrator do? There are several simple ways to solve the problem. The choice depends on the circumstances. In this case, the integrator decided to set up a data mart (that is, a central information repository) where new customer information would get entered.

The data mart would export customer information to Goldmine and Great Plains on a nightly basis. The data mart is a custom solution built on Microsoft SQL Server. It automatically assigns a unique identifier to each new customer thereby eliminating any possible confusion as to which customer is being referenced.

As for the gift retailer, all they needed was a link from the Avexxis system to the website. By creating a simple data mart on the web server which is fed nightly from the Avexxis system, the web server always has up to date information about customers. Now some custom software on the web server can welcome returning customers and offer ideas for new purchases.

Information Is a Terrible Thing to Waste

If you're like most companies, you have a wealth of information about your customers. Probably more than you realize. You need to implement some simple process changes and software solutions to take advantage of that data before it's too late. Yes, too late. Customer information does not improve with age and neither will your business.

We have a project approach that is unique to DAMICON. It enables us to deliver effective results fast. We encourage you to read more about our approach and decide for yourself.

lean production

Ten Rules of Lean Production

1. Eliminate waste

2. Minimize inventory

3. Maximize production flow

4. Pull production from customer demand

5. Meet customer requirements

6. Do it right the first time

7. Empower workers

8. Design for rapid changeover

9. Partner with suppliers

10. Create a continuous improvement culture